Re: Anyone know ancient versions of XLC?
Greetings, Ben. I am replying directly because I do not have a definitive answer. On 2021-04-08 00:32, Ben Huntsman via cctalk wrote: I know this is a strange place to ask, but it's as good a place as any. Anyone on here used IBM's XLC in very old versions? Anyone know what the argument -qdebug=austlib does? I have the docs for IBM C Set++ 3.1 for AIX (1993) and it is not there. As you probably know, the Austin lab was known for HPC s/w so they may have shipped special debug versions. I can't seem to find any documentation that says... It would have been an argument for the compiler shipping with AIX 3.2.5, I believe. Thanks in advance! Have you tried TUHS? Many on that list go far back at IBM and AIX. Sorry, N.
Re: RSX11D disks on EBAY- anyone interested?
Certainly the published interface is constrained by what was "officially" released. Just pondering if there was an internal engineering roadmap from 16->18->22 bits around this time or did things evolve more discretely? I think it evolved. Remember there was one type of core memory for the 11/03 that only worked in a 16 bit box as did the little 4k memory boards that required the 11/03 to do refresh cycles. Maybe the MSV11-CD memory board (had one of those) as well. I think it used "bits 16-17" for the +15,-5 or something like that and plugging an 18 bit card into a slot like that would cause the usual fireworks. But back in the beginning your choice was RX01 floppies (interrupt so they didn't care), or the RKV11. I've got the 1975 or so version of the 11/03 short book, I can take a look. But I remember very clearly the insanity of the RLV11 controller board with the whole BA11-S requirement for the CD bus (which was repurposed as the PMI on the 11/83). Always wondered if you could run an RLV11 in slots 2-3 on a BA23 box. Probably. C
Re: Lisa Source Code
On 4/9/21 6:05 AM, Paul Koning via cctalk wrote: > Linked lists go back a long way. FAT is one, in a sense. DECtape file > systems for DOS (also RSTS) are linked, in the blocks. The same goes for the > CDC 6000 file system (more precisely, it has contiguous blocks within a > track, and links from the last block in the track to the next track number). > > paul > >> On Apr 9, 2021, at 12:27 AM, Richard Sheppard via cctalk >> wrote: >> >> I believe the original Amiga file system also used a linked-list approach. >> That way you could, theoretically, reconstruct a file from any one of it’s >> data blocks. >> Richard The point with the WORM drive setup I described, is that one may recover *every* file ever written, however the name and attributes of the file may not be recoverable, being stored in the tree, which may not be readable. That is, the linked list may be intact, but the tree not so or vice-versa. The advantage is that searching for a given file is faster using a tree than it would be traversing a list. Reading old papers on the subject, it seems that old optical phase-change storage, while permanent may not have been absolutely reliable. I should also mention that although the WORM discs do store quite a bit of data (in my case 470MB per side), they're not very fast. More floppy-disk speed than hard disk. --Chuck
Re: RSX11D disks on EBAY- anyone interested?
On 4/8/21 4:48 PM, Noel Chiappa via cctalk wrote: > From: Jerry Weiss > I always wondered why the RKV11-D was only 16 bit addressable. The manual (EK-RKV11-OP-001) says: "Since the 11/03 BUS structure has no provision for extended addressing, no connection is made to the bus from these [XM] bits on the RKVII-D." (pg. 3-5). Certainly the published interface is constrained by what was "officially" released. Just pondering if there was an internal engineering roadmap from 16->18->22 bits around this time or did things evolve more discretely? > The DEC RK05 disk subsystem cost $10K list circa 1978 (drive, RKV-11D > controller and cabinet), so this wasn't a trivial purchase. Interesting. Where did you see that listed, just out of curiosity? (I looked in the Jan '84 PDP-11 Systems and Options, my copy of which just showed up,but that's too late; I could probably find it if I pawed through all my DEC sales literature, but I'm too lazy... :-). Noel My reference here is "A source handbook for Digital Equipment Corporation LSI-l I-compatible products" https://link.springer.com/content/pdf/10.3758/BF03205335.pdf Jerry
Re: Lisa Source Code
Linked lists go back a long way. FAT is one, in a sense. DECtape file systems for DOS (also RSTS) are linked, in the blocks. The same goes for the CDC 6000 file system (more precisely, it has contiguous blocks within a track, and links from the last block in the track to the next track number). paul > On Apr 9, 2021, at 12:27 AM, Richard Sheppard via cctalk > wrote: > > I believe the original Amiga file system also used a linked-list approach. > That way you could, theoretically, reconstruct a file from any one of it’s > data blocks. > Richard > > Sent from Mail for Windows 10 >
